Cluster Analysis

This repository contains a project focused on performing cluster analysis to group data into meaningful segments. Cluster analysis is a key technique in unsupervised learning, widely used in fields such as marketing, biology, social sciences, and more.

Table of Contents

1. Hierarchical Clustering
2. K-means

Project Overview

This project applies clustering techniques to identify patterns and groupings in the dataset. It covers the following steps:

• Preprocessing data to handle missing values, normalization, and outliers.
• Applying clustering algorithms such as K-Means, Hierarchical Clustering.
• Visualizing the clusters for better interpretability.
• Evaluating the clusters using metrics such as silhouette score.

Prerequisites

To run the project, ensure you have the following installed:

• Python 3.8 or above
• Libraries:
  • NumPy
  • Pandas
  • Scikit-learn
  • Matplotlib
  • Seaborn
  • SciPy (optional, for hierarchical clustering)

Methodology

1. Clustering Algorithms:

   • K-Means: Identify compact spherical clusters.
   • Hierarchical Clustering: Visualize cluster hierarchy with dendrograms.

2. Evaluation Metrics:

   • Dendrogram
   • The Elbow Method
   • Silhouette Analysis

Results

Key findings and visualizations will be added here, such as:

• Cluster centers and sizes.
• Visual representation of clusters in 2D/3D space.

File:

https://github.com/tanhpuh/cluster_analysis/blob/main/Cluster_Analysis.ipynb